Irene V. Blair
University of Colorado, Boulder
Implicit Stereotypes and Prejudice
“Whites today are, in fact, more prejudiced than they are wont to admit, ” concluded Crosby, Bromley, and Saxe (1980, p. 557) nearly two decades ago. With over 80 demonstrations that stereotypes and prejudice can operate without the perceivers' awareness or intention (i.e., implicitly), psychologists today have even more evidence for this conclusion. However, it is no longer a one-liner. With the development of sophisticated experimental techniques, we have obtained an increasingly fine-grained and complex picture of intergroup bias. Moreover, we have come to understand that, in addition to being more prejudiced than people are wont to admit, they may also be more prejudiced than they can admit. 1
The experimental study of implicit intergroup bias began soon after Crosby et al. concluded that Whites had something to hide. By adapting theories and methods from the study of implicit memory and unconscious perception, stereotype and prejudice researchers gained access to a new level of analysis. Particularly influential was the theory of spreading activation (e.g., Collins & Loftus, 1975; Posner & Snyder, 1975). According to this theory, concepts in one's cognitive representation of the world are spatially organized, such that associated concepts (e.g., bread - butter) are closer together than unassociated concepts (e.g., bread - doctor). Because of this arrangement, the activation of one concept causes the activation of nearby, associated concepts, but not distant, unassociated concepts through the process of spreading activation (for discussions of alternative accounts, see Neely, 1991; Ratcliff & McKoon, 1988). As a consequence, the presentation of one concept (e.g., bread) facilitates the subsequent processing of an associated concept (e.g., butter), compared with an unassociated concept (e.g., doctor). Most important, this sequence of events can occur without the perceiver's awareness or intention (i.e., implicitly), and it can be manipulated and studied experimentally.
In one of the first demonstrations of implicit intergroup bias, Gaertner and McLaughlin (1983) showed that participants were faster to identify paired letter strings if they were consistent than inconsistent with the stereotype. Hence, students were faster to identify Whites-ambitious and Blacks-lazy, than Whites-lazy and Blacks-ambitious. A few years later, Devine (1989) demonstrated that even subliminally presented cues could result in intergroup bias. And recently, researchers have measured implicit intergroup bias through neurological and physiological indicators (e.g., Osterhout, Bersick, & McLaughlin, 1997; Vanman, Paul, Ito, & Miller, 1997). For example, Osterhout et al. showed that event-related brain potentials (ERPs) were sensitive to the stereotype consistency of a statement (e.g., “The secretary bought herself a plane ticket” vs. “The secretary bought himself a plane ticket”).
The excitement generated by the availability of methods to study implicit intergroup bias has produced a large and growing literature with strong evidence for the prevalence of such biases in society. Implicit racial and ethnic bias has received the greatest attention, with over 30 studies showing that Whites have relatively strong implicit negative associations with Blacks (or other non-White groups) and positive associations with Whites (e.g., Bargh, Chen, & Burrows, 1996; Chen & Bargh, 1997; Dasgupta, McGhee, Greenwald, & Banaji, 1999; Dasgupta & Greenwald, 2000; Devine, 1989; Dovidio, Evans, & Tyler, 1986; Dovidio, Kawakami, Johnson, Johnson, & Howard, 1997; Fazio & Dunton, 1997; Fazio, Jackson, Dunton, & Williams, 1995; Gaertner & McLaughlin, 1983; Gilbert & Hixon, 1991; Glaser & Banaji, 1999; Greenwald, McGhee, & Schwartz, 1998; Lepore & Brown, 1997; Locke, MacLeod, & Walker, 1994; Moskowitz, Salomon, & Taylor, 2000; Rudman, Ashmore, & Gary, 2000; Sinclair & Kunda, 1999; Spencer, Fein, Wolfe, Fong, & Dunn, 1998; Vanman et al., 1997; Wittenbrink, Judd, & Park, 1997, in press; Wittenbrink, Judd, Park, & Stone, 1997; for evidence of implicit race stereotypes, see Kawakami, Dion, & Dovidio, 1998; Macrae, Bodenhausen, & Milne, 1995; von Hippel, Sekaquaptewa, & Vargas, 1997). Evidence for implicit gender bias has also begun to